132 results about "Dynamic filtering" patented technology

The invention relates to the technical field of automatic production control and industrial data processing, in particular to a PLC-based gypsum-based flame-retardant plate production line control system and method, and the system comprises a data collection module which collects sensor data in real time; the ideal state construction module is used for constructing a pure ideal state vector; the disturbance simulation module is used for generating a theoretical damaged state vector corresponding to a specific fault mode; the difference calculation module is used for generating real and theoretical deviation vectors; the coupling verification control module is used for calculating a similarity value between the actual deviation vector and the theoretical deviation vector; if the similarity value is larger than a preset judgment threshold value, it is judged that a real physical fault exists at present, and a PLC precise compensation instruction is generated and sent to a production line execution mechanism; otherwise, judging that non-physical noise exists at present, keeping the control parameters of the production line unchanged, and starting a dynamic filtering program; according to the method, the problem that physical faults and sensor clutters are difficult to distinguish in a high-noise environment is effectively solved, and missing report is avoided.

The invention relates to the technical field of the Internet of Things, and discloses an animal physiological feature data acquisition system and analysis method based on the Internet of Things, and the method comprises the following steps: cleaning and standardizing original sensor data, including filling missing data, eliminating noise interference through a dynamic filtering technology, unifying the dimension of multi-source data, and analyzing the data; the method comprises the following steps of: acquiring a Beidou time service time reference, eliminating redundant features, interpolating and aligning asynchronous sampling data to the Beidou time service time reference, correcting a time sequence phase difference of physiological and behavior data through a dynamic path matching technology, and correcting GPS / Beidou positioning drift in combination with Kalman filtering. Through multi-source sensing fusion and an edge-fog-cloud cooperative computing architecture, the effective collection rate of animal physiological data is achieved, abnormal response is delayed and compressed, the flexible energy supply technology and dynamic communication optimization are combined, the endurance and communication success rate is maintained in a pasture, and the false alarm rate is reduced compared with a traditional method; common diseases such as abnormal body temperature, digestive system diseases and the like are accurately warned and covered.

The invention relates to the technical field of building information, and provides a knowledge base construction and retrieval method and system based on a multi-source text in the building field, and the method comprises the following steps: a knowledge base construction stage: constructing a multi-dimensional metadata feature vector for a multivariate text based on a standard classification index table; the method comprises the following steps of: converting and segmenting a document, splicing an end clause with all superior title texts by utilizing a context inheritance algorithm to form a text unit with complete semantics, and performing dynamic filtering based on an analyzed query intention and a metadata vector at a user retrieval stage; then, in the screening set, performing fusion calculation on semantic vector similarity, keyword matching degree and authority offset weight based on effectiveness attribute and implementation time, and performing mixed retrieval and reordering on the text units; and finally selecting a text unit according to a sorting result and inputting the text unit into the large language model to generate answers. According to the method, high-precision and high-compliance intelligent retrieval and question answering of building domain knowledge are realized.

The invention provides an NL2SQL database table dynamic screening method and system based on multi-strategy fusion. The method comprises the steps that a preliminary candidate set is generated, a semantic encoder is used for encoding a user natural language problem and table and column information in a database into vectors, the semantic similarity is calculated, a preset number of tables are screened out from a whole database table based on the similarity, and a preliminary candidate table set is formed; performing deep analysis and reordering, performing multi-dimensional analysis on the tables in the preliminary candidate table set, and calculating a graph relationship compactness score and a content correlation score of each table; and performing fusion decision, inputting the semantic similarity score, the graph relationship compactness score and the content correlation score of each table into a fusion model to obtain a final comprehensive score of each table, and screening out a final correlation table set based on the final comprehensive score. The method has high accuracy and high recall rate, the generation speed and response efficiency of the NL2SQL system are improved, meanwhile, the robustness is high, and the practicability and generalization ability of the system are improved.

The invention provides a high-fidelity anti-compression image watermarking method and system based on spectrum-airspace decoupling, and belongs to the field of information security. Firstly, the watermark information is mapped and remodeled; a watermark encoder based on multi-granularity spectrum-spatial domain feature decoupling is constructed, discrete cosine transform is introduced to filter out high-frequency components, Haar wavelet transform is adopted to realize lossless downsampling, a fast Fourier transform dynamic filter is combined to capture global semantic features, and local texture details are combined through multi-scale spatial domain volume accumulation; designing a physical perception and visual self-adaptive dual embedding strategy, and anchoring watermark energy to an anti-compression brightness channel; constructing an anti-attack layer containing differentiable JPEG compression simulation and mixed noise simulation, and participating in network training; and constructing a decoder and designing a loss function to carry out network optimization. According to the method, the robustness of the watermark under strong compression and complex black box attacks is improved, and extremely high visual imperceptibility is realized through physical and visual constraints.

The invention discloses a visual Token compression method and device based on double-path fusion, and the method comprises the steps: obtaining an initial visual Token through employing a visual encoder; obtaining a learnable query vector, and adopting an attention mechanism to enable the learnable query vector to interact with the initial visual Token to obtain a plurality of context Token; analyzing a self-attention map of the initial visual Token in the visual encoder, so as to screen out a plurality of detail Token according to the self-attention map; obtaining a text prompt Token, and calculating a task correlation score of each detail Token by taking the text prompt Token and the plurality of context Token as a query; dynamically filtering or weighting the plurality of details Token according to the task correlation score to obtain a plurality of filtered details Token; the plurality of filtered detail Tokens are spliced with the plurality of context Tokens, so that a compressed visual Token can be obtained; therefore, the key global and local visual information can be intelligently reserved while the high compression ratio is realized and the calculation cost is reduced.

The invention relates to the technical field of safety monitoring of power transmission equipment, in particular to an intelligent early warning nut and a monitoring method thereof.The nut comprises a quantitative anti-interference sensing and dynamic filtering module, a verification parameter is dynamically calculated based on an interference-verification quantitative model by synchronously collecting loosening distance, vibration and electromagnetic interference data, and the verification parameter is obtained; a real loosening event is accurately identified; the environment coupling trend quantitative prediction module is used for fusing multi-environment factor dynamic weighting such as humidity, wind speed and temperature difference after real loosening is judged, and predicting a future loosening distance through an environment-loosening quantitative correlation model; and the scenarized risk grading response module is used for calculating a scenarized risk quantized value R and triggering grading early warning in combination with the prediction result and the structural importance weight of the bolt mounting position, and generating a strategy instruction containing a maintenance priority. According to the invention, the problems of high false alarm rate, incapability of predictive maintenance and short endurance in the prior art are solved, and accurate sensing, intelligent prediction and graded early warning of the bolt loosening state are realized.

The invention relates to the technical field of earthwork measurement, in particular to an accurate earthwork measurement system based on a laser radar technology. The system comprises a mobile scanning acquisition module, a point cloud intelligent processing module and an engineering quantity accurate calculation and visualization module. The mobile scanning acquisition module dynamically acquires original three-dimensional point cloud data of a construction area through a laser radar; the point cloud intelligent processing module separates ground point cloud from the complex scene point cloud with high precision through a dynamic filtering algorithm; and the project amount accurate calculation and visualization module accurately calculates the total amount of filling and excavation through a boundary adaptive calculation method. According to the invention, a dynamic filtering algorithm and a boundary adaptive calculation method fusing geometric features and physical attributes are provided, and full-automatic, high-precision and visual measurement of the earthwork volume is realized.

The invention relates to the technical field of remote sensing image processing, in particular to a raft laver culture area end-to-end extraction method based on frequency dynamic filtering, and the method comprises the steps: obtaining an original remote sensing image of a raft laver culture area, carrying out the preprocessing, constructing a mask data set for high-level semantic supervision, and carrying out the extraction of the original remote sensing image; constructing a hierarchical supervision neural network model, and performing model training on the constructed hierarchical supervision neural network model, including constructing a composite loss function composed of mask loss, gravitational field loss, vertex classification loss and vertex offset loss; and performing optimization processing on a prediction result output by the hierarchical supervision neural network model, and generating a breeding area contour vector result based on the optimized binary mask and vertex coordinates. The method solves the problem that the extraction precision is suddenly reduced in a scene of relatively strong spatial spectrum characteristic heterogeneity of the breeding area in a traditional method, and improves the extraction accuracy of the breeding area. And reliable technical support is provided for large-area and high-precision monitoring of an offshore raft type laver culture area.

The invention discloses a multi-mode adaptive weight SLAM method and system for a construction site. The method comprises the steps that a mapping model is established through sensor joint calibration, laser radar, camera and IMU data are synchronously collected through hardware, and dynamic filtering, distortion compensation and feature extraction are carried out on point clouds and images; the voxel entropy, the angular resolution and the feature density are calculated in real time to evaluate the degradation degree of each mode, and a visual residual error is dynamically adjusted by adopting an adaptive weight and frame skipping mechanism to ensure reasonable information injection in a degradation scene; iMU pre-integration, LiDAR point-surface residual errors and weighted visual reprojection residual errors are sequentially fused in an incremental Kalman sliding window frame, an automatic repositioning strategy is matched, and a high-precision pose and dense color point cloud map is output. According to the method, calibration, synchronization, degradation detection, sliding window optimization, map release and relocation are integrated, and the mapping precision and robustness can be remarkably improved under the complex working conditions of dust shielding, illumination variation, dynamic obstacles and the like.

The invention discloses an underwater salient target detection method and system based on a double-flow fusion network, and belongs to the technical field of computer vision. The method comprises the following steps: respectively extracting multi-scale features of an RGB image and a depth image through a double-flow encoder; in the shallow layer, fusing and enhancing the edge and detail information of the bimodal features through an edge fusion module; in a deep layer, content-adaptive cross-modal semantic fusion is realized in a frequency domain through a dynamic filtering module; fusing the multi-scale features through a cross-layer aggregation decoder to generate a rough saliency map; extracting detail features from the original RGB image through a global detail purification network; and finally, fusing the rough saliency map and the detail features, and outputting an underwater saliency target prediction map. The objective of the invention is to improve the precision and boundary definition of salient target detection in an underwater complex scene.

The application discloses an image snow removal method based on a Transformer modeling and multi-scale dynamic filtering, and comprises the following steps: a data set containing a snow scene image and a corresponding snow-free image is established; an image snow removal network fusing global Transformer modeling and multi-scale dynamic filtering is constructed, and the image snow removal network is denoted as TranFusionNet; the TranFusionNet is trained by using the data set until a preset loss function converges; and a to-be-removed snow image is processed by using the trained image snow removal network to obtain a snow removal result. The application realizes deep fusion of global dependence modeling and local detail enhancement, achieves a good balance between structure consistency maintenance and detail recovery, can effectively remove complex snow particle interference, and reconstructs a clear and natural visual image, thereby providing an efficient solution for the image recovery field.

The invention discloses an electro-gas welding equipment positioning and deviation rectifying system and method based on optimized Kalman filtering, and relates to the technical field of Beidou navigation positioning.The system comprises an original data acquisition module which is deployed on electro-gas welding equipment and used for obtaining original positioning data of the electro-gas welding equipment in real time based on a Beidou navigation chip, the information is uploaded to a background server through a communication network; the filtering deviation correction module is used for receiving the original positioning data and sequentially executing rule judgment and optimized Kalman filtering processing to generate coordinates after deviation correction; the coordinate position marking module is used for marking the corrected coordinates on an electronic map for visual display; according to the method, the problems of positioning drift, matching errors, track jittering and the like of existing GNSS positioning in a complex factory environment are effectively solved by fusing rule judgment and a dynamic filtering technology, finally, the corrected coordinates are marked on an electronic map in real time, and visual monitoring and abnormal behavior early warning are supported.

The invention discloses an event prediction method and system based on a time sequence hypergraph, and belongs to the technical field of event prediction. The method comprises the following steps: acquiring multi-source heterogeneous data of a target region, defining a unified time index, and generating a time sequence feature sequence of each variable of the region through preprocessing; then, identifying a variable causal relationship in the region based on a frequency domain anti-fact condition mutual information algorithm, and fusing the variable causal relationship with a time sequence evolution relationship to construct a time sequence hypergraph structure representing the interior of the region; a dynamic filter is used for filtering the graph, and deep features of the graph are learned by means of a multi-band spectrum gating mechanism, so that internal complex causal and time sequence modes are effectively captured; and finally, performing dichotomy prediction based on the learned graph representation, and outputting the occurrence probability of future events in the region. According to the method, accurate and explainable event prediction is realized, training and prediction do not need to cross regions, data privacy and calculation efficiency are guaranteed, and stronger robustness is shown for specific data distribution change of the regions.

A method of noise reduction, including: collecting configuration data with agents from configuration items, transmitting the collected configuration data to a server for processing, blocking receipt of selected configuration data by a server load protection module to protect the server from overload, storing the configuration data at the server in a storage, processing the configuration data with a batch noise detection module that identifies noise in the configuration data; the batch noise detection module deduces rules based on the identified noise and stores the rules in a data knowledge repository, receiving a user query from a user using a front-end module, analyzing the configuration data with an analytical module to transform the configuration data to be aligned with the user query, responsive to the analysis applying a runtime noise detection module to identify and suppress transformed configuration data to remove irrelevant transformed configuration data.

The invention relates to the technical field of construction site management, in particular to a highway construction site Internet of Things positioning and tracking system which comprises a fundamental frequency calculation module, a parameter generation module, a central frequency generation module, a central frequency calculation module and a central frequency calculation module. The fundamental frequency calculation module collects engine rotating speed and angular speed signals and calculates engine vibration fundamental frequency by combining ignition factors. The dynamic filtering module is used for configuring the weighting coefficient as a difference equation parameter, blocking bandwidth energy through convolution filtering and generating a purification angular velocity; the trajectory tracking module is used for calculating the purification angular velocity and a zero deviation value, calling a Runge-Kutta algorithm for integral generation of a course angle, and generating a position coordinate in combination with a driving speed. According to the method, the vibration fundamental frequency is inversely calculated by collecting the rotating speed of the engine, the blocking bandwidth is dynamically constructed according to the interference frequency, the frequency characteristic is mapped to be a difference equation coefficient, vibration noise is accurately stripped, course drift is eliminated, and it is ensured that the positioning coordinates matched with the actual track are output.

This invention relates to the technical field of industrial wireless communication. Specifically, it provides a duplicate packet filtering system for industrial wireless networks. The system includes: a packet feature extraction module, a hash value calculation module, a dynamic filter module, and a filtering decision and execution module. The packet feature extraction module extracts features from the current wireless packet. The hash value calculation module calculates the corresponding hash value based on the source MAC address of a first terminal. The dynamic filter module queries the hash RAM table based on the hash value to obtain the SA RAM index. The dynamic filter module retrieves the current entry corresponding to the SA RAM based on the SA RAM index. The dynamic filter module determines whether the current wireless packet is a duplicate packet based on the current entry corresponding to the SA RAM. The filtering decision and execution module filters out the current wireless packet when it is a duplicate packet. This invention reduces network resource waste, significantly reduces terminal processing load, and ensures real-time and reliable communication.

The present application relates to the technical field of ocean current monitoring, and discloses a kind of ocean current meter data quality control method and system based on adaptive filtering.The original data containing ocean current direction and speed are obtained, after being decomposed into eastward and northward flow velocity components, frequency spectrum distribution is obtained by Fourier transform, dynamic filtering window is determined by adaptive filtering algorithm, components are preprocessed and feature vector is extracted to calculate flow direction;According to whether the amplitude of the feature vector changes is over the threshold, select Kalman filter to smooth the component or directly obtain the corrected flow direction data;The dynamic trend is extracted by principal component analysis method for the corrected data, and the main direction feature set is obtained;Based on the analysis of the influence of tides and topography, the environmental influence weight is obtained;The filtering parameter update frequency is adjusted by the weight, and the real-time component is filtered twice to output high-precision ocean current direction data.This method can improve the accuracy of ocean current meter data quality control and improve the monitoring accuracy.

The present application relates to a kind of well gas invasion and loss parameter inversion methods based on unscented Kalman filter, belong to petroleum engineering technical field, comprising: step 1: establish overflow wellbore multiphase flow model;Step 2: establish inversion model;Step 3: solve inversion model;Calculate the inversion result of downhole gas invasion and loss parameter.The application method combines wellbore multiphase flow equation with state space model, uses the nonlinear estimation characteristics of UKF algorithm, carries out dynamic filtering and state updating to wellhead measurable data, realizes the real-time inversion of downhole gas volume fraction, gas invasion flow and loss amount.UKF method can accurately capture the nonlinear characteristics of system without linearization, thereby improving the inversion accuracy and calculation stability under complex conditions.Through example verification, this method has good applicability under single reservoir, multiple reservoir and overflow and leakage coexistence conditions, provides reliable boundary conditions and intelligent control basis for well control system.

The invention discloses a video monitoring system for forage feeding in cattle and sheep breeding, relates to the technical field of video monitoring, and can solve the problem of poor overall performance of subsequent feeding monitoring and analysis caused by low accuracy of livestock contour recognition based on video images in an actual breeding scene. Comprising the following steps: acquiring a monitoring video stream of a breeding area and carrying out edge detection to obtain a plurality of edge lines in each frame of image; determining a real contour edge of each target animal according to the plurality of edge lines in each frame of image; according to the real contour edge of each target animal in each frame of image, determining an overall interference evaluation value of the contour of the target animal and a pseudo edge around the real contour edge, and determining a real interference value in each frame of image according to the overall interference evaluation value and the pseudo edge; and according to the real interference value of each frame of image, dynamically adjusting the de-noising filtering kernel parameter of each frame of image, and according to the adjusted dynamic filtering kernel parameter, performing de-noising processing on each frame of image.

The invention relates to a model disaster recovery oriented adaptive feature filtering federated learning method, and aims to solve the fundamental problem that a global model is difficult to meet individual requirements of clients due to heterogeneous client data distribution in model disaster recovery. The core of the method is as follows: a client dynamically filters and fuses feature information from a local personalized model and a frozen global model according to features of each data sample by adopting a self-adaptive gating mechanism based on a one-dimensional convolutional neural network; introducing maximum mean value difference loss to relieve feature distribution offset between clients; before a client uploads update, local and global model parameters are subjected to weighted aggregation through an attention mechanism so as to improve the update quality of the global model. Compared with the prior art, the method has the advantages that the privacy of the client can be kept, the personalized prediction accuracy of the model in a highly heterogeneous data scene is remarkably improved, and the method has higher convergence speed and higher communication efficiency.

The invention discloses a front-end-driven multi-channel video stream dynamic filtering and rendering method and system, belongs to the technical field of computer communication, and aims to solve the technical problem of how to solve member state synchronization abnormity and picture residue caused by network fluctuation and detection delay. Comprising the following steps: establishing a persistent bidirectional connection between a client and a converged communication message service background through a WebSocket protocol; many-to-many real-time communication is realized based on a WebRTC and MCU scheme; in combination with a WebSocket automatic reconnection mechanism and a WebRTC packet loss recovery mechanism, the reliability and stability of connection between the client and the converged communication message service background in a high-concurrency environment are enhanced; a Vue.js framework is used for front-end development, and a componentized and responsive user interface is achieved; front-end calculation is carried out, and signaling-service double-track data difference detection is triggered when members change; and the screen fusion service is requested again, and screen fusion video stream rendering is carried out.

The invention relates to a path planning method, a storage medium, a product, equipment and a vehicle, and the method comprises the steps: carrying out the dynamic filtering of first position data in a path planning result of a previous frame and second position data in an optimal path result of a current frame, and obtaining an inter-frame dynamic filtering result; wherein the inter-frame dynamic filtering result comprises position data; and performing path curvature smooth optimization on the inter-frame dynamic filtering result based on path pure tracking to obtain a path planning result. The smoothness and accuracy of path planning are improved, repeated shaking of the steering wheel is reduced, and driving safety is ensured.

The invention discloses an infrared multi-gas identification method based on a dynamic filter, and belongs to the technical field of infrared spectrum detection and gas sensing, the method comprises the following steps: firstly, obtaining an original response signal sequence of a multi-sampling channel under the condition of no gas, and then constructing an adaptive weight operator for pure gas with different concentrations to suppress low signal-to-noise ratio channel interference; acting a weight operator on a gas response signal to obtain a weighted response, and mapping the signal to a feature space through orthogonal basis function projection transformation to obtain a feature point representing a gas spectral feature; and finally, extracting a feature fingerprint and a vector center of the known gas, executing the same signal processing flow on the to-be-detected mixed gas, and realizing type identification by combining the distribution of the feature points of the mixed gas in the feature space. According to the method, the system integration level is improved by utilizing the tunable characteristic of the dynamic filter, component identification is realized through feature vector modeling, the mixed gas can still be efficiently distinguished under the conditions of low channel number and low signal-to-noise ratio, and the real-time performance and accuracy of detection are both achieved.

Examples provide a system and method for dynamically filtering candidate item identifiers (IDs) from a pool of item IDs in real-time for automatic labeling of images for use as training data used to train computer vision (CV) models. Images of carts are paired with item receipts. Candidate item IDs are extracted from the receipts. Item recognition inference results generated by CV models are used to pair images of individual items with item IDs identifying the item in each item image. As each candidate item ID is assigned to an item image, the item ID is dynamically filtered. Any candidate item IDs remaining after filtering are assigned to any item images failing to pair with an item ID based on the infer results. The results are presented for review and status update via a user interface device for faster and more accurate auto-labeling of training data for CV models.

The invention discloses a text-to-image generation method and system based on multi-expert arrangement. The method comprises the following steps: firstly, analyzing an input text prompt through an intelligent router based on a graph neural network, and accurately activating an optimal expert subset from a heterogeneous model pool by utilizing a correlation score and a self-adaptive dynamic screening mechanism; secondly, active experts are controlled to generate diversified initial sketches in parallel under the minimum denoising budget; then preferentially rearranging the sketch based on a semantic alignment index, and screening out an optimal candidate image; and finally, extracting complementary structure features of the candidate images, and guiding a target diffusion model to perform collaborative fusion by using a dual ControlNet architecture to generate a high-fidelity image. According to the invention, through a cascade arrangement strategy, video memory occupation and reasoning time are substantially reduced, and a high-fidelity generation effect comparable to a large-scale SOTA model is realized.

The invention relates to an automatic code optimization enhancement method based on a large model, which comprises the following steps that a public code snippet data set D is selected, and each code snippet sample contained in the D comprises an original code snippet, an original code snippet review opinion and an optimized code snippet corresponding to the original code snippet; establishing a static filtering strategy and a dynamic filtering strategy, and then adopting the static filtering strategy and the dynamic filtering strategy to carry out data optimization on D in sequence to obtain D '; and constructing a code enhancement model ACREF based on a large language model, and training the ACREF by using D'to obtain a final trained code enhancement model ACREF '. The model disclosed by the invention can more accurately understand and execute core requirements of code optimization, that is, code quality is improved on the premise of keeping semantics and function correctness.

The present application belongs to the technical field of geological disaster monitoring, and particularly relates to a landslide millimeter-level displacement real-time monitoring method based on Beidou No. 3 dual-frequency observation. The method constructs a short baseline network by arranging GNSS monitoring stations in the deformation sensitive area of the landslide and arranging differential reference stations in the stable bedrock area, and synchronously collects Beidou B1C / B2a dual-frequency observation values. The multipath error is suppressed by the dual mechanisms of hardware radius suppression and algorithm correction, the carrier phase observation values are corrected, the localized RTK solution is executed on the embedded platform, the common error is eliminated by combining the wide and narrow lane combination technology and the double difference model, and then the high-precision displacement data is output through the adaptive dynamic filtering based on the PDOP value. The present application realizes millimeter-level monitoring precision, has strong anti-interference ability and environmental adaptability, can accurately capture the early creep of the landslide, and provides reliable support for disaster warning.

The application discloses a kind of man-machine mapping and deft operation method and system based on virtual hand drive, form complete closed loop from hand perception to deft hand control.It includes: image is collected with monocular RGB camera, after normalization etc., 21 three-dimensional coordinates of joint of hand are predicted through multi-task deep neural network, then data is smoothed through dynamic filtering;Particle swarm optimization algorithm is used to fit the shape parameters of MANO hand model;Based on weighted singular value decomposition, the rotation matrix of root joint is calculated, and the global rotation matrix of non-fingertip joint is calculated recursively combined with hand topology structure;Local coordinate system of hand is established, and angle mapping function is designed to convert joint angle into deft hand control quantity;Through double buffering mechanism interpolation processing low frame rate perception result, hybrid control is realized by fusing current data.The scheme improves precision, robustness, adaptability and safety, and provides technical support for efficient and reliable human-computer interaction.